8 Apr
2005
8 Apr
'05
6:57 p.m.
On Fri, 2005-04-08 at 15:47, Lee Revell wrote:
On Fri, 2005-04-08 at 10:39 -0500, Jan Depner wrote:
IIRC part of it was servicing IRIG-B interrupts. The rest was just
serial port interrupts. It was a pretty simple thing otherwise they
would have had to use a hard real-time system of some sort. I tried to
get them to go with Linux but got shot down.
Jan
Please, don't think I was being
disparaging. I am absolutely
thrilled with the kernel work that has been going on. This thing
rocks! The last I heard (from contractors working for us who were doing
soft real-time on - shudder - Windoze) M$ can only get you to about 30
milliseconds.
I didn't think so, I was just pointing out that Linux is quite close to
being able to do hard RT OOTB. Many people don't realize it yet...
Anyway 30ms seems excessive... what kind of soft RT app was that? The
ASIO drivers I used to use go down to 2.6 ms, which doesn't work well, I
have to use ~5ms. But that requires tuning the system carefully.